JP7437919B2 - Judgment system and judgment program - Google Patents

Description

The present invention relates to a determination system and a determination program.

Conventionally, a technique for generating information regarding a user's health has been known (for example, see Patent Document 1). In this conventional technology, health-related information is generated using the imaging results of an infrared camera and a visible light camera.

JP 2018-110013 Publication

However, in the technology of Patent Document 1, the imaging results of an infrared camera and a visible light camera are used to measure mutually different elements, and health-related information is generated based on the measurement results. There was room for improvement from the perspective of information accuracy. That is, for example, the imaging results of an infrared camera were used to measure skin conditions, and the imaging results of a visible light camera were used to measure sagging, swelling, etc., so the accuracy of measuring skin conditions, etc. There was room for improvement in the accuracy of measuring sagging, swelling, etc., and there was room for improvement from the perspective of the accuracy of health-related information.

The present invention was made in view of the above facts, and an object of the present invention is to provide a determination system and a determination program that can improve the accuracy of determining a subject's health condition.

In order to solve the above-mentioned problems and achieve the purpose, the determination system according to claim 1 is a determination system for determining the health condition of a subject, and includes a first acquisition system for acquiring a thermal image of the subject. a second acquisition means for acquiring a visible light image of the subject, based on at least the thermal image acquired by the first acquisition means and the visible light image acquired by the second acquisition means; , a first estimation means for estimating the physiological information of the subject, which is related to the health condition of the subject; determination means for determining the health condition of the subject ; the first estimation means includes a first process for identifying a predetermined region of the subject based on the visible light image acquired by the second acquisition means; , a second process of measuring the surface temperature of the predetermined site identified in the first process based on the thermal image acquired by the first acquisition means; and a second process of measuring the surface temperature of the predetermined site determined in the second process. a third process of estimating the physiological information based on the surface temperature, and the determination system generates correlation information indicating a correlation between the surface temperature information indicating the surface temperature at the predetermined region and the physiological information. storage means for storing, in the third process, the first estimating means is configured to calculate the correlation information between the surface temperature at the predetermined site measured in the second process and the correlation information stored in the storage means. The physiological information is estimated based on .

In the determination system according to claim 2, in the determination system according to claim 1, in the third process, the first estimating means calculates the surface temperature at the predetermined site measured in the second process, and A plurality of types of the physiological information are estimated based on the correlation information stored in the storage means.

The determination system according to a third aspect of the present invention is the determination system according to the first or second aspect, wherein the correlation information is common to the plurality of subjects .

A determination system according to a fourth aspect is the determination system according to the first or second aspect, wherein the correlation information is unique to each of the plurality of subjects .

The determination system according to claim 5 is the determination system according to any one of claims 1 to 4, including a third acquisition means for acquiring weight information of the subject, and a third acquisition means for acquiring body weight information of the subject; a second estimating means for estimating dehydration amount information indicating the amount of dehydration of the subject based on the body weight information, and the determining means includes the dehydration amount information estimated by the second estimating means; The health condition of the subject is determined based on the physiological information estimated by the first estimation means.

The determination system according to claim 6 is the determination system according to any one of claims 1 to 5, wherein the determination system includes a determination system based on the appearance of the subject based on the visible light image acquired by the second acquisition means. identifying means for identifying an external appearance state that is a state of Determine the health status of the person.

The determination system according to claim 7 is the determination system according to any one of claims 1 to 6, comprising: an infrared camera that images the subject and outputs the thermal image of the subject; a visible light camera that images a person and outputs the visible light image of the subject, the infrared camera and the visible light camera mutually capture images of the subject at the same time, and the first acquisition means , the thermal image of the subject taken by the infrared camera is acquired, and the second acquisition means acquires the visible light image of the subject taken by the visible light camera.

The program according to claim 8 is a determination program for determining the health condition of a subject, and includes a computer, a first acquisition means for acquiring a thermal image of the subject, and a visible light image of the subject. and at least the physiological information of the subject based on the thermal image acquired by the first acquisition means and the visible light image acquired by the second acquisition means, a first estimating means for estimating the physiological information related to the health condition of the subject; and a determining means for determining the health state of the subject based on at least the physiological information estimated by the first estimating means. The first estimation means performs a first process of specifying a predetermined region of the subject based on the visible light image acquired by the second acquisition means, and a first process of identifying a predetermined region of the subject based on the visible light image acquired by the second acquisition means; a second process of measuring the surface temperature at the predetermined site identified in the first process based on the image; and estimating the physiological information based on the surface temperature at the predetermined site measured in the second process. a third process of performing a third process of performing the first estimation; In the third process, the means estimates the physiological information based on the surface temperature at the predetermined site measured in the second process and the correlation information stored in the storage means.

According to the determination system according to claim 1 and the determination program according to claim 8, based on the thermal image acquired by the first acquisition means and the visible light image acquired by the second acquisition means, By estimating physiological information related to the health condition of the subject, for example, physiological information can be estimated based on two types of images, a thermal image and a visible light image. It becomes possible to improve the accuracy of determining the health condition of the subject.
Further, a predetermined part of the subject is identified based on the visible light image acquired by the second acquisition means, and the surface temperature of the specified part of the identified subject is determined based on the thermal image acquired by the first acquisition means. By measuring and estimating physiological information based on the measured surface temperature of a predetermined part of the subject, for example, information that is correlated with physiological information can be obtained from measurable parts (for example, the face, etc.). ) is set as a predetermined part, it is possible to improve the accuracy of estimating the physiological information of the subject, and therefore it is possible to improve the accuracy of determining the health condition of the subject. In addition, if the exposed part of the subject (for example, the face) is used as the predetermined part, the surface temperature of the predetermined part can be measured while the subject is wearing clothes, which increases the convenience of the determination system. It is possible to improve the

According to the determination system according to claim 3 , for example , the physiological information of the subject can be estimated using common correlation information, so the health condition of the subject can be determined using a common standard. things become possible.

According to the determination system according to claim 4 , for example , physiological information of a subject can be estimated using correlation information unique to each subject, thereby improving the accuracy of determining the health state of the subject. It becomes possible to do so.

According to the determination system according to claim 5, by estimating the amount of dehydration information indicating the amount of dehydration of the subject, for example, in addition to physiological information, the amount of dehydration information can also be taken into account for determination. It becomes possible to improve the accuracy of determining the health condition of the subject.

According to the determination system according to claim 6, by specifying the appearance state, which is a state judged from the appearance of the subject, based on the visible light image, for example, in addition to physiological information, the appearance state is also considered. Therefore, it is possible to improve the accuracy of determining the health condition of the subject.

According to the determination system according to claim 7, the infrared camera and the visible light camera image the subject at the same time, and the first acquisition means acquires the thermal image of the subject imaged by the infrared camera; 2. By acquiring the visible light image of the subject captured by the visible light camera, the acquisition means can acquire, for example, a thermal image and a visible light image captured at the same timing, so that physiological information of the subject can be obtained. It is possible to improve the estimation accuracy of , and it is possible to improve the accuracy of determining the health condition of the subject.

FIG. 3 is a diagram illustrating an installation state of a determination device. FIG. 3 is a front view of the determination device. FIG. 2 is a block diagram of a determination device. FIG. 3 is a diagram illustrating worker-related information. It is a front view of a worker's face. It is a figure showing an example of correlation. It is a flowchart of determination processing. It is a diagram illustrating a visible light image and a thermal image. It is a diagram illustrating a visible light image and a thermal image. FIG. 3 is a diagram illustrating an installation state of a determination device. FIG. 2 is a block diagram of a weight measuring device and a determining device. It is a figure which illustrated worker weight information. FIG. 2 is a block diagram of a determination device. FIG. 3 is a diagram illustrating worker-related information. It is a figure showing an example of correlation.

DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiments of a determination system and a determination program according to the present invention will be described in detail below with reference to the accompanying drawings. First, [I] the basic concept of the embodiment will be explained, then [II] the specific contents of the embodiment will be explained, and finally [III] modifications to the embodiment will be explained. However, the present invention is not limited to the embodiments.

[I] Basic Concept of Embodiment First, the basic concept of the embodiment will be explained. Embodiments generally relate to a determination system and a determination program.

Here, the "determination system" refers to a system that determines the health condition of a subject, such as a dedicated system for determining the health condition, or a system used for general purposes (for example, a server computer, a personal computer, a tablet computer, etc.). This concept includes systems that are realized by implementing a function for determining the health status of devices (such as terminals). The "determination system" includes, for example, a first acquisition means, a second acquisition means, a first estimation means, and a determination means, and optionally includes a storage means, a third acquisition means, a second estimation means, a specifying means, an infrared ray camera, and a visible light camera.

This "judgment system" can be applied arbitrarily; for example, it can be applied to screening workers at construction sites, or screening users of schools, hospitals, airports, etc., but the following In the illustrated embodiment, a case will be explained in which the present invention is applied to screening for heat stroke among workers at a construction site.

“Target persons” are people or animals whose health status is to be determined, such as workers (also referred to as “workers” as appropriate), facility users, students, patients, medical personnel, and travelers. This is a concept that includes people, etc. "Health condition" is a condition related to health, and includes, for example, whether or not there is a suspicion that you are suffering from heat stroke, or whether or not you are suspected of having any disease such as influenza, etc. It is.

The "first acquisition means" is a means for acquiring a thermal image of the subject, for example, a means for acquiring a thermal image of the subject captured by an infrared camera. An "infrared camera" is a device that images a subject and outputs a thermal image of the subject. A "thermal image" is an image that shows the amount of heat, and is, for example, an image that visualizes the heat distribution output from the object being imaged. By using this "thermal image," it becomes possible to measure the temperature of the object being imaged.

The "second acquisition means" is a means for acquiring a visible light image of the subject, for example, a means for acquiring a visible light image of the subject captured by a visible light camera. A "visible light camera" is a device that images a subject and outputs a visible light image of the subject. A "visible light image" is an image displayed using visible light. Then, by performing known image recognition on this "visible light image", it becomes possible to recognize the object being imaged.

Note that the above-mentioned infrared camera and visible light camera may be configured to image the subject at the same time, or may be configured to capture images of the subject sequentially. Note that "simultaneously capturing images of a subject" is a concept that indicates, for example, that an infrared camera and a visible light camera mutually capture images of a common subject at the same time. Furthermore, "sequentially imaging the subject" means, for example, imaging the subject using either an infrared camera or a visible light camera, and then capturing an image of the subject using the other of the infrared camera or visible light camera. This is a concept that indicates that

The "first estimation means" is based on at least the thermal image acquired by the first acquisition means and the visible light image acquired by the second acquisition means, and is based on the physiological information of the subject and the health of the subject. It is a means of estimating physiological information related to a state. The "first estimating means" is, for example, means for performing the first to third processing.

"Physiological information" is information related to the health condition of the subject, and is a concept that includes, for example, average skin temperature, thigh temperature, density of clothing and dehydration, heart rate, and core temperature. "Average skin temperature" is the average skin temperature of the subject's body (for example, upper body), and is a concept that refers to the average skin temperature at arbitrary positions (for example, 7 places, etc.) on the upper body. be. "Thigh temperature" is the temperature of the thigh, and is a concept that indicates, for example, the temperature of the base of the leg. "Dressed and dehydrated density" is a concept that indicates the density of the amount of water lost by a subject as a result of performing a task. , assuming that hydration is not provided during work, corresponds to the value obtained by dividing the difference in the weight of the subject before and after the work with clothes on) by the product of the time of the work and the weight of the subject. This is the concept of "Heart rate" is a concept that indicates the number of heartbeats in a predetermined period of time (for example, one minute). "Deep temperature" is a concept that indicates the temperature inside the body.

The "first process" performed by the first estimation means is a process of identifying a predetermined part of the subject based on the visible light image acquired by the second acquisition means. The "predetermined part" is a part of the subject's body (including the face), for example, the entire face, a region of the face that includes the nose and mouth, a region that only contains the nose of the face and the nose of the mouth, This concept includes the armpits, knees, elbows, etc.

The "second process" performed by the first estimation means is a process of measuring the surface temperature at the predetermined site identified in the first process based on the thermal image acquired by the first acquisition means.

The "third process" performed by the first estimation means is a process of estimating physiological information based on the surface temperature at the predetermined site measured in the second process. This concept includes a process of estimating physiological information based on the surface temperature of the body and correlation information stored in the storage means. Note that the correlation information will be described later.

The "determination means" is means for determining the health condition of the subject based on at least the physiological information estimated by the first estimation means, and for example, the dehydration amount information estimated by the second estimation means and the first estimation The concept includes means for determining the health condition of the subject based on the physiological information estimated by the means, and also includes a means for determining the health condition of the subject based on the physiological information estimated by the first estimating means and the appearance condition specified by the specifying means. This is a concept that includes means for determining the health condition of a subject.

In other words, the health condition of the subject is determined using at least physiological information, for example, physiological information, dehydration amount information, or appearance state. A case where determination is made using only information will be described, and a case where determination is made using dehydration amount information or appearance state in addition to physiological information will be described in a modified example.

"Storing means" is means for storing correlation information. "Correlation information" is information indicating a correlation between surface temperature information indicating the surface temperature at a predetermined site and physiological information. Note that this "correlation information" may be information that is common to a plurality of subjects, or information that is unique to each of a plurality of subjects.

The "third acquisition means" is means for acquiring weight information indicating the subject's weight.

The "second estimation means" is a means for estimating dehydration amount information indicating the amount of dehydration (amount of water lost) of the subject based on the body weight information acquired by the third acquisition means. Note that the method for estimating the amount of dehydration is arbitrary, but for example, the amount of dehydration may be calculated as the difference in the weight of the subject before and after the work with clothes on, assuming that no hydration is provided during the work. may also be used.

The "identification means" is means for identifying the appearance state, which is a state determined from the subject's appearance, based on the visible light image acquired by the second acquisition means. "Appearance condition" is the condition of the subject judged from the subject's appearance, such as the subject's expression, the degree of shine on the face, the sweat area ratio, the state of armpit sweat, the amount of clothing worn, and the bend of the waist. This is a concept that includes conditions, etc. The "sweat area ratio" is a concept that indicates the ratio of the area of the sweating area in a given part or all of the subject to the area of the subject, for example, the area of the entire face. This is a concept that indicates the ratio of the area of the sweating region of the face to the area.

In the embodiment described below, the "target person" is a worker at a construction site, and the physiological information, dehydration amount information, or physiological information in the appearance condition is used to prevent the worker from suffering from heatstroke. The case of determining whether or not there is a suspicion of being infected will be explained. Furthermore, a case will be described in which average skin temperature, thigh temperature, and dehydration density of clothes are used as "physiological information."

[II] Specific contents of the embodiment Next, specific contents of the embodiment will be described.

(composition)
First, a determination device according to this embodiment will be explained. FIG. 1 is a diagram illustrating an installation state of the determination device, FIG. 2 is a front view of the determination device, and FIG. 3 is a block diagram of the determination device. In addition, in FIGS. 1 and 2, the X-Y-X axes are mutually orthogonal axes, the Z-axis is along the vertical direction, and the Y-axis and the X-axis are along the horizontal direction. explain.

The determination device 1 is a determination system, and is installed, for example, in a rest area at a construction site, as shown in FIG. , 160cm to 170cm, etc.). The determination device 1 includes, for example, the thermography camera 11 shown in FIG. 3, the visible light camera 12, the recording section 13, and the control section 14.

(Configuration - Thermography camera)
The thermographic camera 11 in FIG. 3 is the above-mentioned infrared camera. Although the specific type and configuration of this thermography camera 11 are arbitrary, for example, as shown in FIG. The device is configured to capture and output an image (using different colors that shows the distribution of heat radiated from each location of the imaged object using thermography technology).

(Configuration - visible light camera)
The visible light camera 12 in FIG. 3 is the visible light camera described above. Although the specific type and configuration of this visible light camera 12 are arbitrary, for example, as shown in FIG. , an image showing the appearance of the object being imaged in visible light, which is imaged using a known image sensor or the like, and outputs the image.

Note that the angle of view (range that can be imaged) of this visible light camera 12 and the above-mentioned thermography camera 11 can be set arbitrarily, but for example, if the visible light camera 12 and the above-mentioned thermography camera 11 have a common angle of view, It is assumed that each camera is configured to be able to image the same range.

(Configuration - Recording section)
The recording unit 13 in FIG. 3 is a recording means for recording programs and various data necessary for the operation of the determination device 1. Although the specific type and configuration of this recording unit 13 is arbitrary, it is configured using, for example, an EEPROM, a Flash memory, or the like. However, instead of or in addition to EEPROM and Flash memory, external storage devices such as hard disks, magnetic recording media such as magnetic disks, optical recording media such as DVDs and Blu-ray discs, ROM, USB memory, and SD Any other storage medium can be used, including electrical storage media such as cards.

The recording unit 13 includes, for example, a worker-related information DB 131, in which threshold information is recorded.

(Configuration - Recording Department - Worker related information DB)
The worker-related information DB 131 is a worker-related information storage unit that stores worker-related information, and in particular, a storage unit that stores correlation information. FIG. 4 is a diagram illustrating worker-related information. "Worker related information" is correlation information, and as shown in Figure 4, information corresponding to the item "worker ID", information corresponding to the item "name information", and information corresponding to the item "skin temperature correlation". information,” information corresponding to the item “thigh temperature correlation information,” and information corresponding to the item “clothing/dehydration density correlation information” are mutually associated.

The information corresponding to the item "worker ID" is worker identification information (hereinafter, identification information is also referred to as "ID") that uniquely identifies the worker ("ID001" etc. in FIG. 4). The information corresponding to the item "Name information" is name information that specifies the name of the worker (in FIG. 4, "AA AA", etc., which is described for convenience and is the name of the worker with "ID001").

The information corresponding to the item "Skin temperature correlation information" is skin temperature correlation information indicating the correlation between the surface temperature (surface temperature information) and the average skin temperature (physiological information) at a predetermined site (in FIG. 4, "ID001 (such as "Ts001" which is the skin temperature correlation information of the worker). Note that "Ts001" and the like in FIG. 4 are described here for convenience of explanation, and detailed explanation will be given later. The same applies to thigh temperature correlation information such as "Tf001" and clothing dehydration density correlation information such as "Td001". The information corresponding to the item "thigh temperature correlation information" is thigh temperature correlation information indicating the correlation between the surface temperature (surface temperature information) and the thigh temperature (physiological information) at a predetermined region (in FIG. 4, "ID001") "Tf001" which is the thigh temperature correlation information of the worker). The information corresponding to the item "clothed dehydrated density correlation information" is clothed dehydrated density correlation information that indicates the correlation between the surface temperature (surface temperature information) and the clothed dehydrated density (physiological information) at a predetermined region (in FIG. 4, (such as "Td001" which is clothing/dehydration density correlation information of the worker with "ID001").

FIG. 5 is a front view of the worker's face. As the predetermined region targeted by the surface temperature used in each piece of correlation information in FIG. 4, only one or both of narrow region R1 and wide region R2 shown in FIG. 5 can be used. The "narrow area" R1 is the predetermined area described above, and is an area where only the nose of the face and the nose of the mouth fit, and as shown in FIG. 5, the mouth does not fit around the nose when viewed from the front. It is a rectangular area. Further, the "wide area" R2 is the above-mentioned predetermined part, and is an area of the face that includes the nose and mouth, and as shown in FIG. 5, is a rectangular area that includes the nose and mouth when viewed from the front. , is a region that includes the narrow region R1 and has a larger area than the narrow region R1. In addition, in this embodiment, a case is exemplified in which only the narrow region R1 of the narrow region R1 or the wide region R2 is used as the predetermined region that is the target of the surface temperature used in each correlation information in FIG. explain.

FIG. 6 is a diagram showing an example of correlation. For example, FIG. 6 is a diagram showing the correlation for the worker with "ID001" in FIG. 4, and the average skin temperature correlation line L1 in FIG. The thigh temperature correlation straight line L2 in FIG. 6(b) shows the correlation between the surface temperature of the narrow region R1 and the thigh temperature, and FIG. 6(c) shows the correlation with the temperature. The clothes dehydrated density correlation line L3 shows the correlation between the surface temperature of the narrow region R1 and the clothes dehydrated density. As "Ts001" which is the skin temperature correlation information in FIG. 4 described above, information indicating the average skin temperature correlation straight line L1 in FIG. 6(a) (for example, a calculation formula or table information) is stored. Furthermore, as "Tf001" which is the thigh temperature correlation information in FIG. 4, information indicating the thigh temperature correlation straight line L2 in FIG. 6(b) (for example, an arithmetic expression or table information) is stored. Further, as "Td001" which is the clothing dehydration density correlation information in FIG. 4, information indicating the clothing dehydration density correlation straight line L3 in FIG. 6(c) (for example, a calculation formula or table information, etc.) is stored. shall be. Note that each correlation information of workers other than the worker of "ID001" in FIG. Assume that information is stored.

The specific storage method for the worker-related information shown in FIG. ) is communicatively connected to the determination device 1, and when each piece of information is input to the determination device 1 via the terminal, the control unit 14 stores the input information based on the input information. Regarding the skin temperature correlation information, the manager measures the average skin temperature of each worker using any measuring device, and also measures the surface of the narrow area R1 of the worker at the time of measuring the average skin temperature. Measure the temperature, input a large number of combinations of the average skin temperature and the surface temperature of the narrow area R1 (for example, 30 to 50 sets) into your own terminal, and then input them into your own terminal. An average skin temperature correlation straight line is obtained by performing arbitrary processing (for example, processing for obtaining an approximate straight line) on this large amount of data, and information indicating the obtained average skin temperature correlation straight line is input into the determination device 1. In this case, the control unit 14 stores the information based on each input information. Note that the thigh temperature correlation information and the clothing/dehydration density correlation information are also stored in the same manner as the skin temperature correlation information. Note that the storage method described here is an example, and the storage method is not limited to this method, and other methods may be used.

(Configuration - Recording section - Threshold information)
The "threshold information" stored in the recording unit 13 in FIG. 3 is information indicating a threshold for determining whether or not a worker is suspected of suffering from heat stroke. This information indicates a threshold value set for each piece of physiological information (specifically, average skin temperature, thigh temperature, and dehydration density of clothing) whose correlation is shown in each piece of correlation information. Note that this threshold information may be set individually for each worker or may be set commonly for all workers, but in this embodiment, the threshold information is set for all workers. An example of a common setting will be explained.

Here, for example, based on academic papers, experiments, simulations, etc. related to heat stroke, the threshold value for average skin temperature (hereinafter also referred to as "average skin temperature threshold value") is "35.5 (℃)", the thigh temperature "34.5 (℃)" which is the threshold value for "thigh temperature threshold" (hereinafter also referred to as "thigh temperature threshold value"), and "7.5 (g /kgh)'' is stored as threshold information.

(Configuration - control section)
The control unit 14 is a control unit that controls the determination device 1, and specifically includes a CPU, various programs interpreted and executed on the CPU (basic control programs such as an OS, and specific functions activated on the OS). A computer is equipped with an internal memory such as a RAM for storing programs and various data.

Functionally, the control unit 14 includes, for example, a first acquisition unit 141, a second acquisition unit 142, a first estimation unit 143, and a determination unit 144. The first acquisition unit 141 is a first acquisition unit that acquires a thermal image of the subject. The second acquisition unit 142 is a second acquisition unit that acquires a visible light image of the subject. The first estimation unit 143 calculates physiological information of the subject based on at least the thermal image acquired by the first acquisition unit 141 and the visible light image acquired by the second acquisition unit 142. This is a first estimation means for estimating physiological information related to a state. The determining unit 144 is a determining unit that determines the health condition of the subject (worker) based on at least the physiological information estimated by the first estimating unit 143. The processing of each part of the control section 14 will be described later.
(Determination process)
Next, the determination process executed by the determination device 1 configured as described above will be explained. FIG. 7 is a flowchart of the determination process (steps are abbreviated as "S" in the description of each process below). The "determination process" is roughly a process executed by the determination device 1, and is, for example, a process of determining whether or not a worker is suspected of suffering from heat stroke. The timing of executing this determination process is arbitrary, but for example, the execution is started when the power of the determination device 1 is turned on and the thermography camera 11 and the visible light camera 12 are started. I will explain from where it started. Here, for example, a worker at a construction site is placed at a predetermined position in front of the determination device 1 in the rest area shown in FIG. The explanation will be given assuming that the user visits the location) and receives the determination by the determination device 1. Further, each worker at the construction site will be described below on the assumption that the worker-related information shown in FIG. 4 is stored in advance in the worker-related information DB 131. Further, a case will be described by way of example in which a determination is made regarding the worker with "ID001" in FIG.

At SA1 in FIG. 7, the control unit 14 determines whether or not to acquire an image. Although the specifics are arbitrary, for example, when the determination process is activated, the control unit 14 uses the thermography camera 11 and the visible light camera 12 to capture an image at a predetermined angle of view (range that can be captured) of each camera. It is assumed that the configuration is such that images are repeatedly captured at time intervals (for example, at intervals of 0.5 seconds to 1 second, etc.) and the captured images are received. More specifically, the control unit 14 is configured to simultaneously capture images at predetermined time intervals with the thermography camera 11 and the visible light camera 12, and receive the thermal image and visible light image captured at the same timing. That's it. Then, the control unit 14 determines whether the received thermal image and visible light image are images of the worker's face (that is, whether the worker's face is captured in each image), Based on the determination result, it is determined whether or not to acquire an image.

The specific method for determining whether the received thermal image and visible light image are images of the worker's face is arbitrary, but for example, a method using known image recognition may be used. . More specifically, for example, a visible light image that shows the worker's face and a visible light image that does not show the face are used to perform machine learning with teacher information to create a visible light image that shows the worker. Model information for image recognition (that is, when a visible light image is input, model information that outputs information indicating whether or not a worker is included in the input visible light image) is generated in advance. A case will be described in which a determination is made using the generated model information. Note that the determination method here is an example, and other methods may be used. For example, when a thermal image is input, information indicating whether or not a worker is included in the input thermal image is output. The model information may be used, or the determination may be made using another method without using the respective model information.

Specifically regarding SA1, the control unit 14 first receives a thermal image and a visible light image output from the thermography camera 11 and the visible light camera 12 (that is, a thermal image and a visible light image captured at the same timing). , input the received visible light image into the model information described above. Next, if "information indicating that the worker is not included in the input visible light image" is output from the model information, the control unit 14 determines that the received thermal image and visible light image do not show the worker's face. After determining that the image has not been captured, it is determined that the image is not to be acquired (NO in SA1), and SA1 is repeatedly executed until it is determined that the image is to be acquired. In addition, if "information indicating that a worker is included in the input visible light image" is output from the model information, the control unit 14 detects that the received thermal image and visible light image capture the face of the worker. After determining that the image has been acquired, it is determined that the image is to be acquired (YES in SA1), and the process moves to SA2.

FIG. 8 is a diagram illustrating a visible light image and a thermal image. Note that the thermal image is actually an image that shows the distribution of heat radiated from each position of the imaged object using different colors, but in Figure 8, the distribution of heat is shown for convenience of explanation. The entire image is illustrated by hatching without specifically illustrating the difference in color. The same applies to illustrations corresponding to thermal images in other figures. Here, for example, when the worker with "ID001" visits a predetermined position in front of the determination device 1 in the rest area of FIG. The thermal image and visible light image shown in FIG. 8 are received as the images. Then, when the acquired visible light image of FIG. 8 is input into the above-mentioned model information, "information indicating that the worker is included in the input visible light image" is output from the model information, and the received thermal image is After determining that the visible light image is a captured image of the worker's face, it is determined that the image is to be acquired.

At SA2 in FIG. 7, the control unit 14 acquires an image. Specifically, this is optional, but for example, the first acquisition unit 141 acquires a thermal image among the thermal images and visible light images determined to be output in SA1, and the second acquisition unit 142 acquires a visible light image of the thermal image and visible light image determined to be output in SA1. Here, for example, the first acquisition unit 141 acquires the thermal image shown in FIG. 8, and the second acquisition unit 142 acquires the visible light image shown in FIG.

In SA3 of FIG. 7, the first estimating unit 143 identifies a region corresponding to the face and a region corresponding to a predetermined part in the thermal image acquired in SA2. Although the specifics are optional, for example, by performing image recognition on the visible light image acquired by the second acquisition unit 142 in SA2, the outline of the worker's face, nose, and mouth can be identified in the visible light image. Furthermore, in the visible light image, a rectangular narrow image region GR1 that does not include the mouth and is centered around the position of the nose is specified as a region corresponding to a predetermined region (hereinafter referred to as "first processing of SA3"). Note that the "narrow image region" GR1 is a region on the image corresponding to the aforementioned narrow region R1. Next, a reference point is set at a predetermined position (for example, the upper left corner point) in the visible light image acquired by the second acquisition unit 142 in SA2, and the contour of the identified worker's face and The position or range of the region corresponding to the specified predetermined region is specified by any method (for example, a method of specifying it as a set of coordinates based on a reference point, etc.) (hereinafter referred to as "second processing of SA3"). Next, a predetermined position in the thermal image acquired by the first acquisition unit 141 in SA2 (a predetermined position in the thermal image corresponding to the position where the reference point was set in the visible light image described above, for example, a point in the upper left corner) A reference point is set in the thermal image, and based on the set reference point and the position or range of the region corresponding to the outline and predetermined part of the worker's face identified in the "second processing of SA3" described above, In this step, the outline of the worker's face and the position or range of the region corresponding to the predetermined region are identified.

Note that the image recognition performed in the "first processing of SA3" can use known image recognition technology that recognizes the face itself or the nose and mouthpiece of the face from a visible light image, so a detailed explanation will be given below. Omitted.

FIG. 9 is a diagram illustrating a visible light image and a thermal image. Here, for example, in the "first process of SA3", by performing image recognition on the visible light image of FIG. The nose and mouth are identified, and a rectangular narrow image region GR1 (FIG. 9(a)) that does not include the mouth is centered around the position of the nose in the visible light image, and is located at a predetermined location (that is, the narrow region R1 in FIG. 5). ). Next, in the "second processing of SA3", a predetermined position (for example, the upper left corner point) in the visible light image of FIG. 8 acquired by the second acquisition unit 142 in SA2 is set as shown in FIG. A reference point is set, and the contour of the worker's face specified above and the position or range of the specified narrow image region GR1 are determined using any method (for example, specifying the position or range of the specified narrow image region GR1 as a set of coordinates based on the reference point). method, etc.). Next, in the "third process of SA3", a predetermined position in the thermal image of FIG. As shown in FIG. 9(b), a reference point is set at the position (for example, the point in the upper left corner), and the set reference point and the face of the worker identified in the "second processing of SA3" described above are Based on the contour of the worker's face and the position or range of the narrow image region GR1, the contour of the worker's face and the position or range of the narrow image region GR1 are specified on the thermal image. Then, as shown in FIGS. 9(b) and 9(c), it is possible to specify the position or range of the area corresponding to the face and the narrow image area GR1 in the thermal image. It becomes possible to specify the distribution and the heat distribution of the narrow region R1 (that is, a predetermined region) in FIG. 5, and it becomes possible to measure the surface temperature of the region corresponding to the face and the surface temperature of the narrow region R1. Note that the "first processing of SA3" here corresponds to the "first processing" performed by the first estimating means.

At SA4 in FIG. 7, the first estimation unit 143 measures the surface temperature of a predetermined portion. Although the specifics are arbitrary, for example, based on the color in the area corresponding to the predetermined region identified in SA3 in the thermal image acquired in SA2 (that is, the color indicating the heat distribution in the thermal image), the predetermined Measure the surface temperature of the area. Note that the processing here corresponds to the "second processing" performed by the first estimating means.

Here, for example, the color (that is, the heat distribution in the thermal image) in the narrow image region GR1 in FIG. 9(b) or FIG. The surface temperature of the narrow region R1 in FIG. 5 is measured based on the color shown in FIG. More specifically, for example, information indicating the relationship between a color in a thermal image and a temperature expected from the heat indicated by the color (for example, a calculation formula or table information, etc.) is recorded in the recording unit 13, With reference to this information, the temperature distribution corresponding to the color distribution of each pixel (each pixel) in the narrow image region GR1 of FIG. 9(b) or FIG. 9(c) is specified, and the temperature distribution of each specified pixel is The average temperature is calculated, and the calculated average temperature is taken as the measurement result of the surface temperature of the narrow region R1 in FIG. Here, for example, "Tm" is used as the surface temperature of the narrow region R1 in FIG. ) will be measured.

At SA5 in FIG. 7, the first estimation unit 143 estimates physiological information. Specifically, it is optional, but first, by performing image recognition on the visible light image acquired in SA2, the work whose information is stored in the worker-related information in FIG. 4 of the worker-related information DB 131 in FIG. 3 is performed. The operator identifies which worker among the workers is shown in the above-mentioned visible light image (hereinafter referred to as "first process of SA5"). Note that image recognition here is optional, but for example, machine learning with teacher information can be performed using visible light images of each worker whose information is stored in the worker-related information in Figure 4. (In other words, if a visible light image is input, model information that identifies the worker in the input visible light image (for example, worker ID or name) A case will be described in which model information) for outputting information) is generated in advance and identification is performed using the generated model information. Next, with reference to the worker-related information in FIG. 4 in the worker-related information DB 131 in FIG. Correlation information (specifically, skin temperature correlation information, thigh temperature correlation information, clothing dehydration density correlation information) is acquired (hereinafter referred to as "second processing of SA5"). Next, the surface temperature of the predetermined region measured in SA4 is obtained, and based on each correlation information obtained in the "second processing of SA5", physiological information corresponding to the obtained surface temperature of the predetermined region is identified. , the identified physiological information is estimated as the physiological information of the worker in the visible light image. Note that the processing here corresponds to the "third processing" performed by the first estimating means.

Here, for example, first, in the "first process of SA5", by performing image recognition on the visible light image of FIG. 8 acquired in SA2, the worker-related information of FIG. 4 of the worker-related information DB 131 of FIG. It is specified which worker among the workers whose information is stored in the above-mentioned visible light image. Specifically, the visible light image in FIG. 8 is input to the model information described above, and in this case, for example, "worker ID" = "ID001" is output from the model information, and "ID001" in FIG. The worker identifies the person shown in the visible light image of FIG. Next, in the "second process of SA5", the worker-related information in FIG. 4 of the worker-related information DB 131 in FIG. The correlation information associated with the worker "ID001" in FIG. Obtain “Td001”. Next, in the "third process of SA5", "Tm" is obtained as the surface temperature of the predetermined part measured in SA4, and based on each correlation information obtained in the "second process of SA5", the obtained predetermined temperature is Physiological information corresponding to "Tm", which is the surface temperature of the site, is "average skin temperature" = "V1" in FIG. "Thigh temperature" = "V2" in FIG. 6(b) (that is, the value specified as the thigh temperature corresponding to "Tm" in the thigh temperature correlation line L2), and FIG. 6( c) "Clothed dehydrated density" = "V3" (that is, the value specified as the clothed dehydrated density corresponding to "Tm" in the clothed dehydrated density correlation line L3), and these identified physiological information are shown in the figure. This is estimated as the physiological information of the worker with "ID001" shown in the visible light image of No.8. Note that "V1" to "V3" here are for convenience.

At SA6 in FIG. 7, the determining unit 144 determines whether there is a suspicion that the worker is suffering from heatstroke. Specifically, it is optional, but for example, after acquiring the threshold information recorded in the recording unit 13 and also acquiring the physiological information estimated in SA5, comparing the acquired threshold information and the physiological information, Based on the comparison results, it is determined whether the worker is suspected of suffering from heatstroke.

In detail, first, the average skin temperature estimated in SA5 and the "average skin temperature threshold" indicated by the threshold information are compared with each other, and the estimated average skin temperature is estimated to be equal to or higher than the "average skin temperature threshold". It is determined whether the average skin temperature is less than the "average skin temperature threshold" (hereinafter referred to as "first determination of SA6"). In addition, the thigh temperature estimated in SA5 is compared with the "thigh temperature threshold" indicated by the threshold information, and whether the estimated thigh temperature is equal to or higher than the "thigh temperature threshold" is determined. (hereinafter referred to as "second determination of SA6"). In addition, the clothed dehydrated density estimated in SA5 is compared with the “clothed dehydrated density threshold” indicated by the threshold information, and whether the estimated clothed dehydrated density is greater than or equal to the “clothed dehydrated density threshold” or the estimated clothed dehydrated density is It is determined whether it is less than the "clothed dehydrated density threshold" (hereinafter referred to as "third determination of SA6").

Then, based on the determination results of these "first determination of SA6" to "third determination of SA6", it is determined whether or not the worker is suspected of suffering from heatstroke. Although the specifics are optional, for example, if there is one or more judgment results that the estimated physiological information is equal to or higher than the threshold indicated by the threshold information, there is a suspicion that the worker is suffering from heat stroke. On the other hand, if all the estimated physiological information is less than the threshold indicated by the threshold information (in other words, if there is no judgment result that the estimated physiological information is greater than or equal to the threshold indicated by the threshold information), the worker It is determined that there is no suspicion that the person is suffering from heat stroke.

Here, for example, in the threshold information, as described above, "average skin temperature threshold" = "35.5 (℃)", "thigh temperature threshold" = "34.5", and "clothed dehydration density threshold" = "7.5 (g/kgh)" is stored, the above-mentioned "V1" is less than "35.5 (℃)", the above-mentioned "V2" is "34.5" or more, and the above-mentioned If "V3" of It is determined that the patient is suspected of suffering from heat stroke.

At SA7 in FIG. 7, the control unit 14 outputs the determination result at SA6. Although the specifics are optional, for example, it is assumed that a display device (not shown) is provided in a position visible to the worker around the determination device 1 in FIG. 1, and information indicating the determination result of SA6 is displayed on the display device. The information may also be transmitted to and displayed on the display device. Alternatively, for example, with reference to the worker-related information in FIG. Alternatively, the identified name information and information indicating the determination result of SA6 may be transmitted to the administrator's terminal device and displayed on the terminal device. Alternatively, this information may be stored in the recording section 13. Or, only when it is determined in SA6 that the person is suspected of suffering from heatstroke, guidance information indicating methods to avoid heatstroke (for example, a guidance message saying "Please drink water frequently") may be displayed on a display device. Note that a speaker device may be provided around the determination device 1 in FIG. 1 or in the determination device 1 itself, and the determination result or guidance information may be outputted as a voice through the speaker device. This completes the determination process.

(Effects of this embodiment)
According to the present embodiment, based on the thermal image acquired by the first acquisition unit 141 and the visible light image acquired by the second acquisition unit 142, physiological information of the worker who is the target person is acquired. By estimating physiological information related to a person's health condition, for example, it is possible to estimate physiological information based on two types of images, a thermal image and a visible light image, which improves the accuracy of determining the health condition of a person. It is possible to improve the

Further, based on the visible light image acquired by the second acquisition unit 142, a predetermined part of the subject is specified, and based on the thermal image acquired by the first acquisition unit 141, the specified part of the subject is By measuring the surface temperature and estimating physiological information based on the measured surface temperature at a predetermined part of the subject, for example, information that is correlated with the physiological information can be obtained from measurable parts (e.g. When the predetermined part is a face, etc.), the accuracy of estimating physiological information of the subject can be improved, and therefore the accuracy of determining the health condition of the subject can be improved. Furthermore, when the exposed part of the subject (for example, the face, etc.) is set as the predetermined part, the surface temperature of the predetermined part can be measured while the subject is wearing clothes, making the determination device 1 convenient. It is possible to improve sexuality.

In addition, correlation information indicating the correlation between surface temperature information indicating the surface temperature at a predetermined region and physiological information is stored, and the correlation information is unique to each of multiple subjects, so that, for example, Since the physiological information of the subject can be estimated using correlation information unique to each subject, it is possible to improve the accuracy of determining the health status of the subject.

Further, the thermography camera 11 and the visible light camera 12 mutually image the subject at the same time, the first acquisition unit 141 acquires the thermal image of the subject imaged by the thermography camera 11, and the second acquisition unit 142 By acquiring the visible light image of the subject captured by the visible light camera 12, it is possible to acquire a thermal image and a visible light image captured at the same timing, thereby improving the accuracy of estimating the subject's physiological information. This makes it possible to improve the accuracy of determining the health condition of the subject.

[III] Modifications to the Embodiments The embodiments of the present invention have been described above, but the specific configuration and means of the present invention are within the scope of the technical idea of each invention described in the claims. can be arbitrarily modified and improved. Hereinafter, such a modified example will be explained.

(About the problems to be solved and the effects of the invention)
First of all, the problems to be solved by the invention and the effects of the invention are not limited to the above-mentioned contents, but may differ depending on the implementation environment of the invention and the details of the configuration, and only some of the problems described above can be solved. In some cases, the above-mentioned effects may be solved or only some of the above-mentioned effects may be achieved.

(About distribution and integration)
Further, each of the electrical components described above is functionally conceptual, and does not necessarily need to be physically configured as shown in the drawings. In other words, the specific form of distribution and integration of each part is not limited to what is shown in the diagram, and all or part of it may be functionally or physically distributed or integrated in arbitrary units depending on various loads and usage conditions. can be configured. Furthermore, the term "system" in this application is not limited to one configured by a plurality of devices, but includes one configured by a single device. Furthermore, the term "device" in this application is not limited to one configured by a single device, but includes one configured by a plurality of devices.

(About shape, numbers, structure, time series)
With respect to the components illustrated in the embodiments and drawings, the shape, numerical values, structure or chronological relationship of multiple components may be arbitrarily modified and improved within the scope of the technical idea of the present invention. Can be done.

(About physiological information)
Furthermore, in the embodiment described above, a case has been described in which skin temperature, thigh temperature, and dehydration density of clothes are used as physiological information, but the invention is not limited to this, and for example, information on only one of these pieces of information, or It may be configured to use two pieces of information.

(Regarding the area corresponding to the face)
Further, in the above embodiment, a case has been described in which the area corresponding to the face is specified in SA3, but the invention is not limited to this. For example, only the area corresponding to a predetermined part is specified without specifying the area corresponding to the face. It may be configured to do so.

(About worker-related information (Part 1))
Moreover, the storage method of the skin temperature correlation information, the thigh temperature correlation information, and the clothing dehydration density correlation information in the worker-related information in FIG. 4 of the above embodiment may be arbitrarily changed. For example, to measure the surface temperature of the worker's narrow area R1, the control unit 14 performs a process similar to the process of SA1 to SA4 in FIG. Each correlation information may be generated and stored using a combination of R1's surface temperature, average skin temperature, thigh temperature, and clothing dehydration density input by the administrator.

(About worker-related information (Part 2))
Further, in the above embodiment, the skin temperature correlation information, thigh temperature correlation information, and clothing dehydration density correlation information in FIG. 4 are unique to each worker, but the present invention is not limited to this, for example. , may be common to each worker. In this case, for example, the surface temperature of the narrow region R1, the average skin temperature, the thigh temperature, and the dehydration density of clothes are measured for a large number of workers, and these averages are used to determine the temperature that is commonly used by a large number of workers. It is also possible to generate and store skin temperature correlation information, thigh temperature correlation information, and clothing dehydration density correlation information one by one, and to use and process each of these correlation information in common for a large number of workers. good. When configured in this way, correlation information indicating the correlation between surface temperature information indicating the surface temperature at a predetermined region and physiological information is stored, and the correlation information is common to multiple subjects. By doing this, for example, the physiological information of the subject can be estimated using common correlation information, so that the health condition of the subject can be determined using a common standard.

(About dehydration amount information)
Further, in the above embodiment, information on the amount of dehydration of the worker may also be taken into consideration to determine whether or not the worker is suspected of suffering from heat stroke. FIG. 10 is a diagram illustrating an installation state of the determination device, and FIG. 11 is a block diagram of the weight measuring device and the determination device. Note that in FIG. 11, the same components as those in FIG. 3 are designated by the same reference numerals, and their explanations will be omitted. Although the specific implementation method is arbitrary, for example, as shown in FIG. 10, a weight measuring device 3 that can communicate with the determining device 2 is provided.

(About dehydration amount information - Configuration - Judgment device)
The determination device 2 is a determination system, and is the same as the determination device 1 in FIG. It is something.

The communication unit 20 is a communication means for communicating with an external device (weight measurement device 3), and can be configured using, for example, a known communication circuit.

The recording unit 23 includes the worker-related information DB 131 and the worker weight information DB 232 described in the embodiment, and also records threshold information.

The worker weight information DB 232 is a worker weight information storage unit that stores worker weight information. FIG. 12 is a diagram illustrating worker weight information. "Worker weight information" is information indicating the weight of the worker before and after work, and as shown in Figure 12, information corresponding to the item "worker ID" and information corresponding to the item "name information". , information corresponding to the item "weight information before work", and information corresponding to the item "weight information after work" are mutually associated. The information corresponding to the item "worker ID" and the item "name information" is the same as the information with the same name in FIG. 4. The information corresponding to the item "Pre-work weight information" is pre-work weight information indicating the weight before performing the work (in FIG. 12, "W11" indicates the value of the pre-work weight of the worker with "ID001"). etc). Note that "W11" shown here is a description for convenience (the same applies to "W12" described later). The information corresponding to the item "weight information after work" is the weight information after work indicating the weight after work (in FIG. 12, "W12" indicating the value of the weight of the worker with "ID001" after work) "etc). Although the specific storage method for such worker weight information is arbitrary, for example, the worker ID and name information may be stored by copying the information with the same name in FIG. 4, and A method of storing the pre-work weight information and post-work weight information will be described later.

The "threshold information" stored in the recording unit 23 of FIG. 11 is roughly the same as the "threshold information" stored in the recording unit 13 of FIG. 3, and includes, for example, the above-mentioned average skin temperature threshold, In addition to the thigh temperature threshold and the clothing dehydration density threshold, it is assumed that "1.5 (kg)" is stored as a dehydration amount threshold that is a threshold related to the dehydration amount.

Functionally, the control unit 24 in FIG. 11 includes, for example, the first acquisition unit 141, the second acquisition unit 142, and the first estimation unit 143 described in the embodiment, as well as the third acquisition unit 245 and the first estimation unit 143. A second estimating section 246 is provided, and the determining section 144 is replaced by a determining section 244. The third acquisition unit 245 is a third acquisition unit that acquires weight information indicating the weight of the worker who is the subject. The second estimating unit 246 is a second estimating unit that estimates dehydration amount information indicating the amount of dehydration of the subject based on the weight information acquired by the third acquiring unit 245. The determining unit 244 is a determining unit that determines the health condition of the subject based on the dehydration amount information estimated by the second estimating unit 246 and the physiological information estimated by the first estimating unit 143. The processing of each part of the control section 24 will be described later.

(About dehydration amount information - Configuration - Weight measurement device)
The weight measuring device 3 shown in FIGS. 10 and 11 is a measuring device that measures the weight of the worker placed on the weight measuring device 3, and can be constructed using, for example, a known weighing scale technique. This weight measuring device 3 is used by a worker at a construction site at a predetermined position in front of the determining device 2 in the rest area shown in FIG. It is provided on the front side of the determination device 2 so that the weight of the worker can be measured when the worker visits the location.

(About dehydration amount information - processing)

The determination device 2 performs the same processing as the determination device 1 unless otherwise specified, and only the parts that are different from the processing of the determination device 1 will be described. Here, we will explain a case where, for example, before and after work, a worker visits a predetermined position in front of the determination device 2 to take an image and measure his/her weight, as shown in FIG. 10. .

When the worker visits a predetermined position in front of the determination device 2 before work as shown in FIG. Communication is carried out, weight information indicating the weight of the worker measured by the weight measuring device 3 is obtained, and the worker is identified by performing the same process as "SA5 first process" in FIG. 7. In the worker weight information of FIG. 12 of the worker weight information DB 232, the acquired weight information described above is stored as the pre-work weight information associated with the identified worker. Note that, here, the third acquisition unit 245 determines whether the current time is before or after the work, for example, if the determination device 2 is provided with a timer (such as a timer) and can specify the current time. If the current time falls within the time period before the start of work (for example, 8:45 a.m. to 9:00 a.m. and 12:45 p.m. to 1:00 p.m.), , and if it falls within the time period after the work is completed (for example, 12:00 p.m. to 12:15 p.m., and 17:00 p.m. to 17:15 p.m., etc.) It may be identified as

Here, for example, if the worker with “ID001” visits a predetermined position in front of the determination device 2 as shown in FIG. communicates with the weight measuring device 3 to obtain weight information “W11” indicating the weight of the worker measured by the weight measuring device 3, and in the worker weight information in FIG. "W11" is stored as the pre-work weight information associated with "ID"="ID001".

Next, when the worker comes to a predetermined position in front of the determination device 2 after the work, as shown in FIG. 10, the third acquisition unit 245 of the determination device 2 communicates Communication is performed with the weight measuring device 3 via the weight measuring device 3, and weight information indicating the weight of the worker measured by the weight measuring device 3 is obtained, and the same process as the “first process of SA5” in FIG. 7 is performed. The process is performed to identify the worker, and in the worker weight information in FIG. 12 of the worker weight information DB 232 in FIG. 11, the acquired weight information is used as post-work weight information associated with the identified worker. Store. Here, for example, when the worker with "ID001" comes to a predetermined position in front of the determination device 2 after work, as shown in FIG. Similarly, communication is performed with the weight measuring device 3 via the communication unit 20, and "W12" which is the weight information indicating the weight of the worker measured by the weight measuring device 3 is obtained. In the worker weight information, "W12" is stored as the post-work weight information associated with "worker ID"="ID001".

Next, the second estimating unit 246 refers to the worker weight information in FIG. 12 and acquires the above-mentioned stored post-work weight information and pre-work weight information of the same worker as the post-work weight information. Then, the difference between the obtained post-work weight information and the obtained pre-work weight information is calculated, and the calculated difference is estimated as the amount of dehydration of the worker. Here, for example, the second estimating unit 246 refers to the worker weight information in FIG. "Pre-work weight information" = "W11" is obtained, the difference between the obtained "W11" and the obtained "W12" is calculated, and the calculated difference is estimated as the amount of dehydration of the worker with "ID001".

Next, the determining unit 244 determines whether the worker is suffering from heat stroke based on the amount of dehydration estimated by the second estimating unit 246 and the physiological information estimated by the first estimating unit 143 by performing the same processing as in the embodiment. Determine whether or not there is a suspicion that the patient is suffering from the disease. Although the specifics are arbitrary, for example, in addition to the "first determination of SA6" to "third determination of SA6" described in the embodiment, the amount of dehydration estimated by the second estimation unit 246 and the The dehydration amount threshold indicated by the threshold information recorded in the recording unit 23 of 11 is compared with each other to determine whether the estimated dehydration amount is greater than or equal to the "average dehydration amount threshold" or the estimated dehydration amount is equal to the "dehydration amount threshold". (hereinafter referred to as "dehydration amount determination").

For example, if there is one or more determination results that the estimated physiological information is equal to or higher than the threshold value indicated by the threshold information (as in the case of the embodiment), or if the estimated dehydration amount is "average dehydration amount" If it is determined that the worker is at or above the "threshold value", it is determined that the worker is suspected of suffering from heat stroke, and on the other hand (as in the case of the embodiment), all the estimated physiological information is If it is determined that the estimated amount of dehydration is less than the threshold and the estimated amount of dehydration is less than the "average dehydration amount threshold," it is determined that there is no suspicion that the worker is suffering from heat stroke.

Here, for example, if the difference between "W12" and "W11" as the dehydration amount of the worker with "ID001" is equal to or more than "1.5 (kg)" as the dehydration amount threshold, the determination unit 244 It was determined that the person was suspected of suffering from heatstroke.

With this configuration, by estimating the amount of dehydration information that indicates the amount of dehydration of the subject, for example, in addition to physiological information, it is possible to take into account the information on the amount of dehydration, so that the health state of the subject can be determined. It becomes possible to improve the determination accuracy.

(About appearance condition)
Further, in the embodiment described above, it may be configured to determine whether or not there is a suspicion that the worker is suffering from heat stroke, taking into consideration the appearance state of the worker. FIG. 13 is a block diagram of the determination device. Note that in FIG. 13, the same components as those in FIG. 3 are designated by the same reference numerals, and the description thereof will be omitted.

(About appearance condition - configuration)
The determination device 4 is a determination system, and is the same as the determination device 1 shown in FIG.

The recording unit 43 includes a worker-related information DB 431 and records the threshold information described in the embodiment.

The worker-related information DB 431 is a worker-related information storage unit that stores worker-related information, and in particular, a storage unit that stores correlation information. FIG. 14 is a diagram illustrating worker-related information. “Worker related information” is correlation information, and as shown in FIG. 14, information corresponding to the item “worker ID”, information corresponding to the item “name information”, and information corresponding to the item “skin temperature correlation Information corresponding to the item ``Thigh Temperature Correlation Information'', information corresponding to the item ``Clothing and Dehydration Density Correlation Information'', and information corresponding to the item ``Sweat Area Ratio Correlation Information'' are mutually correlated. Associated. Note that information other than the information corresponding to the item "sweat area ratio correlation information" is the same as the information with the same name in FIG. 4. The information corresponding to the item "Sweat area ratio correlation information" is sweat area ratio correlation information indicating the correlation between the surface temperature and the sweat area ratio (appearance condition) at a predetermined region (in FIG. 14, the information corresponding to "ID001" ``Ta001'', which is worker sweat area ratio correlation information, etc.).

FIG. 15 is a diagram showing an example of correlation. Note that FIG. 15 is a diagram showing the correlation for the worker with "ID001" in FIG. 14, and the sweat area ratio correlation straight line L4 shows the correlation between the surface temperature of the narrow region R1 and the sweat area ratio. There is. The sweat area ratio here will be explained below as, for example, indicating the ratio of the area of the sweating region of the worker's face to the area of the worker's face. As "Ta001" which is the sweat area ratio correlation information in FIG. 14 described above, information indicating the sweat area ratio correlation straight line L4 in FIG. 15 (for example, an arithmetic expression or table information) is stored. shall be. The specific storage method for the worker-related information in FIG. 14 is arbitrary; for example, it is assumed that the worker-related information in FIG. 14 is stored in advance in the same manner as the worker-related information in FIG. 4 described in the embodiment. . In particular, the sweat area ratio correlation information is stored based on information on the health status of the worker when there is no suspicion of heatstroke. Regarding the sweat area ratio here, we used the ratio of the area of the sweating area on the face to the area of the face that includes the narrow area R1, but this is not limited to the area that does not include the narrow area R1. The ratio of the area of the published area in the relevant part to the area of the area may be used.

Functionally, the control unit 44 in FIG. 13 includes, for example, a specifying unit 445 in addition to the first acquisition unit 141, second acquisition unit 142, and first estimation unit 143 described in the embodiment, The determination unit 144 is replaced by a determination unit 444. The specifying unit 445 is a specifying unit that specifies the appearance state, which is a state determined from the subject's appearance, based on the visible light image acquired by the second acquisition unit 142. The determining unit 444 is a determining unit that determines the health condition of the subject based on the external appearance specified by the specifying unit 445 and the physiological information estimated by the first estimating unit 143. The processing of each part of the control section 44 will be described later.

(About appearance condition - treatment)

The determination device 4 performs the same processing as the determination device 1 unless otherwise specified, and only the portions that are different from the processing of the determination device 1 will be described.

In SA5 of FIG. 7, in addition to the process described in the embodiment, the identification unit 445 identifies the sweat area ratio of the worker based on the visible light image acquired by the second acquisition unit 142. Although the specifics are optional, for example, the second acquisition unit 142 acquires the visible light image in SA2, performs the same process as in the "first process of SA3", and then Identify the contours of the worker's face. Next, in the visible light image mentioned above, image recognition is further performed in the area inside the identified facial contour (in other words, the area corresponding to the face) to identify the sweating area, and Distinguish between sweating and non-sweating areas. Note that the method of recognizing the sweating area here is arbitrary, but for example, an image recognition method may be used that focuses on the fact that the sweating area has shine due to sweat, or , we performed supervised machine learning using images of faces where part or all of the face is sweating, and images of faces that are not sweating to determine the sweating area of the worker in the visible light image. Model information for specifying may be generated in advance, and specification may be performed using the generated model information. Next, we calculate the ratio of the area identified as sweating to the total area inside the facial contour (i.e., the area of the entire face) (i.e., the area identified as sweating). ÷ "area of the entire area inside the contour of the face"), and the calculation result is specified as the sweat area ratio of the worker.

At SA6 in FIG. 7, the determination unit 444 determines whether the worker Determine whether or not there is a suspicion that the person is suffering from heat stroke. Although specific details are arbitrary, for example, in addition to the "first determination of SA6" to "third determination of SA6" described in the embodiment, the following determination is performed. Specifically, the surface temperature of a predetermined part measured in SA4 and the sweat area ratio identified by the aforementioned identifying unit 445 are acquired, and the sweat area ratio identified by the aforementioned identifying unit 445 is acquired, and the sweat area ratio that is shown in the visualized image of the worker-related information in FIG. 14 is obtained. After acquiring the sweat area ratio correlation information associated with the worker, the information on the combination of the acquired surface temperature of the predetermined site and the sweat area ratio is determined by the sweat area ratio correlation straight line L4 indicated by the acquired sweat area ratio correlation information. It is determined whether the deviation exceeds a predetermined percentage. More specifically, in FIG. 15, it is determined whether the point specified by the surface temperature and sweat area ratio of the predetermined region described above deviates from the region between the upper limit straight line L41 and the lower limit straight line L42.

For example, if there is one or more determination results that the estimated physiological information is equal to or higher than the threshold indicated by the threshold information (as in the case of the embodiment), or if the surface temperature and sweat area ratio of a predetermined region are If the combination of information deviates from the sweat area ratio correlation straight line L4 indicated by the sweat area ratio correlation information by a predetermined percentage or more, it is determined that the worker is suspected of suffering from heatstroke; (Similar to the case of the form), all the estimated physiological information is less than the threshold value indicated by the threshold value information, and the information on the combination of the surface temperature and sweat area ratio of the predetermined region is equal to or less than the sweat area indicated by the sweat area ratio correlation information. If the ratio does not deviate from the correlation line L4 by a predetermined percentage or more, it is determined that there is no suspicion that the worker is suffering from heatstroke.

Here, for example, if the surface temperature of the predetermined region measured by SA4 is "34.0 (°C)" and the sweat area ratio identified by the identification unit 445 is "70 (%)", the surface temperature of the predetermined region is Since the information on the combination of temperature and sweat area ratio corresponds to point P1 in FIG. 15, the information on the combination of surface temperature and sweat area ratio of a predetermined region corresponds to the sweat area ratio correlation indicated by the sweat area ratio correlation information. Since it deviates from the straight line L4 by a predetermined percentage or more, it is determined that the worker is suspected of suffering from heatstroke.

When configured in this way, by specifying the appearance state, which is a state judged from the appearance of the subject, based on the visible light image, for example, in addition to physiological information, the appearance state can also be considered for determination. Therefore, it is possible to improve the accuracy of determining the health condition of the subject.

(About a method for identifying the area corresponding to the face and the area corresponding to a predetermined part in a thermal image)
Furthermore, in the above embodiment, a case has been described in which a region corresponding to a face and a region corresponding to a predetermined part in a thermal image are specified by setting the reference point in FIG. 9 at SA3 in FIG. do not have. For example, using AI-related technology (e.g., machine learning technology), the area corresponding to the face and the area corresponding to a predetermined part in the visible light image, the area corresponding to the face and the area corresponding to the predetermined part in the thermal image, The configuration may be such that a region corresponding to the face and a region corresponding to a predetermined region in the thermal image are specified by superimposing the two regions.

(About other features)
Further, in the above embodiments or modified examples, the configuration may be changed as appropriate. Specifically, for example, dehydration symptoms, core body temperature, and body surface temperature are predicted by measuring the surface temperature of the worker's face based on the imaging results of the thermography camera 11 and the visible light camera 12, and the weight of only workers with severe symptoms is predicted. Measurements and body temperature measurements may be performed in detail, reducing the time and burden required for monitoring by workers.

Further, for example, the determination device 1 may be of a battery-driven type or a power source (100V commercial power source or solar power source) driven type.

Further, for example, the thermography camera 11 of the determination device 1 may be used to monitor suspicious persons at night, and the determination device 1 may be provided with a light emitting unit (configured to output flash light) to output the flash light. It may be configured such that a visible light image of a suspicious person is captured by capturing an image with the visible light camera 12 at a certain timing. With this configuration, the determination device 1 can also be used for crime prevention purposes.

Further, for example, the health condition determined by the determination device 1 may be changed depending on the season, and for example, a configuration may be configured such that a determination regarding heat stroke is performed in the summer, and a determination regarding influenza is performed in the winter.

For example, based on the image captured by the visible light camera 12 of the determination device 1, the state of the person (expression, degree of shine on the face, state of sweating from underarm sweating, amount of clothing, or curve of the user's waist) may be determined. heatstroke may be determined based on the identification result.

Further, for example, the components of the determination device 1 may be arranged in a distributed manner. For example, the thermography camera 11 and the visible light camera 12 may be provided separately from the determination device 1, and each of these cameras and the determination device 1 may be connected to each other. It may be configured to perform the processing described in the embodiment by communicating.

Further, for example, a waterproof and dustproof casing may be used as the housing of the determination device 1, and the functions of the determination device 1 may be implemented in a wearable device (for example, a glasses-type device, etc.). It may be mounted on a robot, or it may be linked to a facial recognition system when entering and exiting the venue.

(About combinations)
Further, the techniques of the above embodiments and modifications may be combined arbitrarily. In particular, by combining the techniques "(About dehydration amount information)" and "(About appearance state)" of the embodiment and the modification, the health of workers is determined by taking into account physiological information, dehydration amount information, and appearance information. It may be configured to determine the state.

(Additional note)
The determination system of Supplementary note 1 is a determination system for determining the health condition of a subject, and includes a first acquisition unit that acquires a thermal image of the subject, and a second acquisition unit that acquires a visible light image of the subject. and physiological information of the subject, based on at least the thermal image acquired by the first acquisition means and the visible light image acquired by the second acquisition means, the health condition of the subject. and a determination means that determines the health condition of the subject based on at least the physiological information estimated by the first estimation means.

The determination system according to appendix 2 is the determination system according to appendix 1, in which the first estimation means specifies a predetermined region of the subject based on the visible light image acquired by the second acquisition means. a second process of measuring the surface temperature at the predetermined site identified in the first process based on the thermal image acquired by the first acquisition means; A third process of estimating the physiological information based on the surface temperature at the site is performed.

The determination system according to Supplementary note 3 is the determination system according to Supplementary note 2, further comprising: storage means for storing correlation information indicating a correlation between surface temperature information indicating the surface temperature at the predetermined site and the physiological information; The correlation information is common to the plurality of subjects, and in the third process, the first estimation means stores the surface temperature at the predetermined region measured in the second process and the storage means. The physiological information is estimated based on the stored correlation information.

The determination system according to Supplementary Note 4 is the determination system according to Supplementary Note 2, further comprising: storage means for storing correlation information indicating a correlation between surface temperature information indicating the surface temperature at the predetermined site and the physiological information; The correlation information is unique to each of the plurality of subjects, and in the third process, the first estimation means calculates the surface temperature at the predetermined site measured in the second process and the storage means. The physiological information is estimated based on the correlation information stored in the physiological information.

The determination system according to Supplementary note 5 is the determination system according to any one of Supplementary notes 1 to 4, which includes: a third acquisition means for acquiring weight information of the subject; and a third acquisition means for acquiring body weight information of the subject; a second estimating means for estimating dehydration amount information indicating the amount of dehydration of the subject based on the dehydration amount information of the subject, and the determining means includes the dehydration amount information estimated by the second estimating means and the first estimating means. The health condition of the subject is determined based on the physiological information estimated by.

The determination system according to appendix 6 is the determination system according to any one of appendices 1 to 5, in which the state is determined based on the appearance of the subject based on the visible light image acquired by the second acquisition means. specifying means for specifying a certain appearance state, and the determining means determines the health state of the subject based on the appearance state specified by the specifying means and the physiological information estimated by the first estimating means. Determine.

The determination system according to appendix 7 is the determination system according to any one of appendices 1 to 6, which includes: an infrared camera that images the subject and outputs the thermal image of the subject; a visible light camera that outputs the visible light image of the subject, the infrared camera and the visible light camera mutually capture images of the subject at the same time, and the first acquisition means outputs the visible light image of the subject. The second acquiring means acquires the visible light image of the subject photographed by the visible light camera.

The program in appendix 8 is a determination program for determining the health condition of a subject, and includes a computer as a first acquisition unit that acquires a thermal image of the subject, and a second acquisition unit that acquires a visible light image of the subject. an acquisition means, based on at least the thermal image acquired by the first acquisition means and the visible light image acquired by the second acquisition means, physiological information of the subject; It functions as a first estimating means for estimating the physiological information related to the health condition, and a determining means for determining the health condition of the subject based on at least the physiological information estimated by the first estimating means.

(Effect of appendix)
According to the determination system described in Appendix 1 and the determination program described in Appendix 8, the physiology of the subject is determined based on the thermal image acquired by the first acquisition means and the visible light image acquired by the second acquisition means. By estimating physiological information related to the subject's health condition, for example, it is possible to estimate physiological information based on two types of images, a thermal image and a visible light image. It becomes possible to improve the accuracy of determining the health status of people.

According to the determination system described in Supplementary Note 2, a predetermined region of the subject is identified based on the visible light image acquired by the second acquisition means, and a predetermined region of the subject is identified based on the thermal image acquired by the first acquisition means. By measuring the surface temperature of a predetermined part of a subject and estimating physiological information based on the measured surface temperature of a predetermined part of the subject, for example, information that is correlated with the physiological information can be obtained. When a measurable part (for example, a face, etc.) is set as a predetermined part, it is possible to improve the accuracy of estimating the physiological information of the subject, and therefore it is possible to improve the accuracy of determining the health condition of the subject. In addition, if the exposed part of the subject (for example, the face) is used as the predetermined part, the surface temperature of the predetermined part can be measured while the subject is wearing clothes, which increases the convenience of the determination system. It is possible to improve the

According to the determination system described in Appendix 3, correlation information indicating the correlation between surface temperature information indicating the surface temperature at a predetermined region and physiological information is stored, and the correlation information is common to multiple subjects. By doing so, for example, the physiological information of the subject can be estimated using common correlation information, so that the health condition of the subject can be determined using a common standard.

According to the determination system described in Appendix 4, correlation information indicating a correlation between surface temperature information indicating the surface temperature at a predetermined region and physiological information is stored, and the correlation information is stored for each of a plurality of subjects. By being unique, for example, the physiological information of a subject can be estimated using correlation information unique to each subject, making it possible to improve the accuracy of determining the health status of the subject. Become.

According to the determination system described in Appendix 5, by estimating the amount of dehydration information indicating the amount of dehydration of the subject, for example, in addition to physiological information, it is possible to make a determination considering the amount of dehydration information. It becomes possible to improve the accuracy of determining a person's health condition.

According to the determination system described in Appendix 6, by specifying the appearance state, which is a state determined from the appearance of the subject, based on the visible light image, for example, the appearance state is taken into consideration in addition to physiological information. Therefore, it is possible to improve the accuracy of determining the health condition of the subject.

According to the determination system described in Appendix 7, the infrared camera and the visible light camera image the subject at the same time, and the first acquisition means acquires the thermal image of the subject imaged by the infrared camera; By acquiring the visible light image of the subject captured by the visible light camera, the acquisition means can acquire, for example, a thermal image and a visible light image captured at the same timing, so that physiological information of the subject can be obtained. Estimation accuracy can be improved, and it becomes possible to improve determination accuracy of the subject's health condition.

1 Determination device 2 Determination device 3 Weight measurement device 4 Determination device 11 Thermography camera 12 Visible light camera 13 Recording section 14 Control section 20 Communication section 23 Recording section 24 Control section 43 Recording section 44 Control section 131 Worker related information DB
141 First acquisition unit 142 Second acquisition unit 143 First estimation unit 144 Determination unit 232 Worker weight information DB
244 Judgment unit 245 Third acquisition unit 246 Second estimation unit 431 Worker related information DB
444 Judgment unit 445 Specification unit GR1 Narrow image area L1 Average skin temperature correlation line L2 Thigh temperature correlation line L3 Clothing/dehydration density correlation line L4 Sweat area ratio correlation line L41 Upper limit line L42 Lower limit line P1 Point R1 Narrow area R2 Wide area

Claims (8)

A determination system for determining the health condition of a subject,
a first acquisition means for acquiring a thermal image of the subject;
a second acquisition means for acquiring a visible light image of the subject;
Based on at least the thermal image acquired by the first acquisition means and the visible light image acquired by the second acquisition means, physiological information of the subject, which is related to the health condition of the subject. first estimation means for estimating the physiological information;
determination means for determining the health condition of the subject based on at least the physiological information estimated by the first estimation means ;
The first estimating means is
a first process of identifying a predetermined part of the subject based on the visible light image acquired by the second acquisition means;
a second process of measuring the surface temperature at the predetermined site identified in the first process based on the thermal image acquired by the first acquisition means;
performing a third process of estimating the physiological information based on the surface temperature at the predetermined site measured in the second process;
The determination system includes:
Storage means for storing correlation information indicating a correlation between surface temperature information indicating a surface temperature at the predetermined region and the physiological information,
The first estimating means is
In the third process, the physiological information is estimated based on the surface temperature at the predetermined site measured in the second process and the correlation information stored in the storage means.
Judgment system. The first estimating means is
In the third process, a plurality of types of physiological information are estimated based on the surface temperature at the predetermined site measured in the second process and the correlation information stored in the storage means.
The determination system according to claim 1. The correlation information is common to the plurality of subjects ,
The determination system according to claim 1 or 2 . The correlation information is unique to each of the plurality of subjects ,
The determination system according to claim 1 or 2 . a third acquisition means for acquiring weight information of the subject;
a second estimation means for estimating dehydration amount information indicating the amount of dehydration of the subject based on the body weight information acquired by the third acquisition means;
The determining means determines the health condition of the subject based on the dehydration amount information estimated by the second estimating means and the physiological information estimated by the first estimating means.
The determination system according to any one of claims 1 to 4. A specifying means for specifying an appearance state that is a state determined from the appearance of the subject based on the visible light image obtained by the second obtaining means,
The determining means determines the health condition of the subject based on the appearance state specified by the specifying means and the physiological information estimated by the first estimating means.
The determination system according to any one of claims 1 to 5. an infrared camera that images the subject and outputs the thermal image of the subject;
a visible light camera that images the subject and outputs the visible light image of the subject;
the infrared camera and the visible light camera mutually image the subject at the same time;
The first acquisition means acquires the thermal image of the subject captured by the infrared camera,
The second acquisition means acquires the visible light image of the subject captured by the visible light camera.
The determination system according to any one of claims 1 to 6. A determination program for determining the health condition of a target person,
computer,
a first acquisition means for acquiring a thermal image of the subject;
a second acquisition means for acquiring a visible light image of the subject;
Based on at least the thermal image acquired by the first acquisition means and the visible light image acquired by the second acquisition means, physiological information of the subject, which is related to the health condition of the subject. first estimation means for estimating the physiological information;
functioning as a determining means for determining the health condition of the subject based on at least the physiological information estimated by the first estimating means ;
The first estimating means is
a first process of identifying a predetermined part of the subject based on the visible light image acquired by the second acquisition means;
a second process of measuring the surface temperature at the predetermined site identified in the first process based on the thermal image acquired by the first acquisition means;
performing a third process of estimating the physiological information based on the surface temperature at the predetermined site measured in the second process;
The computer includes:
Storage means for storing correlation information indicating a correlation between surface temperature information indicating a surface temperature at the predetermined region and the physiological information,
The first estimating means is
In the third process, the physiological information is estimated based on the surface temperature at the predetermined site measured in the second process and the correlation information stored in the storage means.
Judgment program.

Images